Who will win the Super Bowl? It's a question that every sports talking head is currently trying to answer. In that spirit, I'm going to add to the noise and use some historical data to narrow our search down a bit.
There have been twenty Super Bowls since the league expanded to 32 teams in the 2002 season. I thought it would be interesting and instructive to evaluate those twenty winners on how they stacked up leaguewide in as many important statistics as I could think of, with an eye on finding patterns that might show what team traits were most important. I looked at 22 different measures (points scored, points allowed, yards gained and allowed, passing yards gained and allowed, rushing yards gained and allowed, ANY/A for both offense and defense, passer rating for offense and defense, yards per rush gained and allowed, sack % created and allowed, 3rd down % converted and allowed, red zone TD % scored and allowed, turnover margin, and special teams DVOA) and recorded where each champion ranked. To decide which of these criteria provided meaningful predictive value, I disregarded each statistical measure in which the eventual champion did NOT finish in the top ten (or better) at least 70% of the time. For example, our twenty champions only finished in the top ten in special teams eight times in 20 (40%). While I believe that special teams are important, it's clear that having a top shelf unit is not critical to ultimate success.
Having done all this, I've determined seven key statistical markers to sort out the contenders from the pretenders:
Points are king. This really is common sense but worth noting. 75% of all champions finished in the top ten in scoring. More notably, 70% finished in the top eight of scoring defense. These are the first two of our tests.
Passing matters, but not volume as much as efficiency. Interestingly, while neither passing yards gained or passing yards allowed achieved the sought-after statistical significance, both of the passing efficiency statistics did for both offense and defense. In fact, 70% of eventual winners were in both the top ten of offensive passer rating AND ANY/A. The exact same distinction applies to the defensive side; 70% of winners were in the top ten in both defensive statistics. Being in the top ten in both measures represents tests #3 and #4.
Rushing doesn't really matter, except in one way. While neither rushing yards gained or rushing yards surrendered proved statistically significant, rushing yards surrendered per attempt did: 70% of eventual winners ranked in the top ten in this defensive category.
Arguably the most telling stat of all is, unsurprisingly, turnovers. No particular surprise here, but Test #6 tells us that 75% of eventual winners placed in the top eight in turnover margin.
Finally, the stats in aggregation matter as well. As mentioned before, while most of the stats I mentioned failed to achieve the significance I was looking for, I did note that most winners were broadly proficient in the vast majority of these measures (this makes sense, all of these are important things for teams to be good at); 70% of all winners averaged 12th or better leaguewide across all 22 measures.
To sum up, here are the seven tests:
TEST 1: top ten scoring offense
TEST 2: top eight scoring defense
TEST 3: top ten in both offensive ANY/A and passer rating
TEST 4: top ten in both defensive ANY/A and passer rating
TEST 5: top ten in yards per rush surrendered
TEST 6: top eight in turnover margin
TEST 7: averages 12th or higher leaguewide in the 22 stats measured
Of the twenty Super Bowl winners I looked at, the average number of "passed tests" was five. Eleven (55%) passed either six or all seven. Only one (the 2007 Giants) passed fewer than three. It seems eminently reasonable that we should be focusing on teams that pass at least five of the tests, and we can ignore those that pass fewer than three. Here's how the 14 playoff teams did:
Passed all seven: 49ers, Bengals
Passed six: Eagles
Passed five: Cowboys, Bills
Passed four: Jaguars
Passed three: Chiefs
Passed fewer than three (discarded "contenders"): Buccaneers, Chargers, Dolphins, Giants, Ravens, Seahawks, Vikings. Neither the Giants nor the Bucs passed a single test.
Click below to see a table showing the NFL ranks in these stats for the fourteen playoff teams:
A few observations/conclusions:
I imagine the big surprises would be how well the Bengals and Jaguars did, and how weakly the Chiefs scored. Frankly, the first round bye that KC secured is likely the biggest boost to their odds that one can find; not only is their defense likely too weak to be a major threat, but more ominously, NONE of the twenty prior winners included a current season MVP winner (Patrick Mahomes is heavily favored to win the award). This should make any Eagles fans who are smarting over Hurts losing out on the award feel a little better.
Cincinnati and Jacksonville feature bright young QBs leading explosive offenses, underrated defenses and both carry momentum into the playoffs. I like both to win this weekend.
By this methodology, the worst game of the weekend should be Minnesota-New York. Eagles fans should really be rooting for the Giants to pull this off; it will all but guarantee that Philly will get the weakest possible team that advances as far as the divisional round, as this honor will in all probability go to the winner of this game.
Miami really tested poorly through this analysis, and that was WITH their top two QBs. They'll be starting rookie third string QB Skylar Thompson this weekend against a serious Buffalo team. That game will almost certainly be the NFL playoff version of Georgia-TCU.
Wierdly, the Dolphins are not the only team starting a rookie 3rd string QB in these playoffs. The 49ers' Brock Purdy makes his playoff debut against Seattle this weekend, and SF will be heavily (and justifiably) favored. San Francisco clearly poses the greatest threat to Philly in the NFC, and who would be favored in a hypothetical matchup between the two will come down to Philly's guaranteed home field edge, and who is healthy for the game.
Dallas grades out pretty well by this methodology, but for lots of reasons I have no confidence in them (Three playoff wins in 25 years? No road playoff wins in 30 years? Got boat raced by a Commander team with nothing to play for last week? Where do you hang your hat, if you're looking for confidence?). Still, Tampa doesn't look good at all and they are (along with Jacksonville) one of two home underdogs this weekend.
For the reasons listed above, I rate the 49ers, Bengals, Bills and Eagles as the serious contenders. The Chiefs, Cowboys and Jaguars are possibilities but not on the same level. The remaining teams will ultimately end up in the "happy to be here" basket.
In the end, what does all this mean for the Eagles' chances? Frankly, they look pretty good. Philly passes more tests than Buffalo, they have a better defense and OL than Cincinnati, and they'll have a guaranteed home game against SF. I would definitely like their chances in a rubber match against Dallas at home. No one else in the bracket is really well-rounded enough to pose a serious threat.
Of course, as far as the Eagles go, their good health is the paramount thing. Health questions surround Jalen Hurts, Lane Johnson, Josh Sweat, Avonte Maddox, Miles Sanders and CJGJ. Who knows how many of those guys will be available, and how close to 100% they'll be? We'll know more in seven days.
Comments